ClawHub
Back to skill

Security audit

AI KEY·标题与封面

Security checks for vulnerabilities and agentic risk

Overview

This is a disclosed title-and-cover writing helper, with no executable payload or hidden destructive behavior, though it may use local AI KEY memory files if available.

Install this if you want a structured Chinese title and cover-text workflow. Be aware that it may read AI KEY reference/vault files and save title feedback or performance notes locally, so review those memory writes if you do not want persistent style preferences.

A.I.G

Vulnerability Patterns
  • Skill Instruction HijackingAlters the agent's session goals or safety constraints when the skill loads
  • Agent Memory PoisoningWrites attacker-controlled rules into memory that affect later sessions
  • Remote Payload Retrieval and ExecutionFetches external code whose behavior can change after review
  • Embedded Malicious CodeShips malicious scripts inside the skill and executes them locally
  • Unauthorized Access and Privilege EscalationObtains permissions beyond the task's legitimate needs

SkillSpector

Vulnerability Patterns
  • Data ExfiltrationExternal Transmission, Env Variable Harvesting, File System Enumeration
  • Trigger AbuseOverly Broad Trigger, Shadow Command Trigger, Keyword Baiting Trigger
  • Prompt InjectionInstruction Override, Hidden Instructions, Exfiltration Commands
  • Privilege EscalationExcessive Permissions, Sudo/Root Execution, Credential Access
  • Supply ChainUnpinned Dependencies, External Script Fetching, Obfuscated Code
Findings (2)

Vague Triggers

Medium
Confidence
95% confidence
Finding
The trigger list includes broad natural-language phrases such as requests for titles or cover text that users may say in ordinary conversation, which can cause accidental activation. Unintended invocation can expose unrelated user content to this skill’s strong behavioral constraints and file-reading workflow, creating context hijacking and confusing or undesired outputs.

Natural-Language Policy Violations

Medium
Confidence
86% confidence
Finding
The skill enforces a specific language and style policy ('direct judgment', 'no jargon', and similar constraints) by default rather than treating it as a preference the user can accept or override. While not directly enabling code execution or data exfiltration, this can override user intent, reduce transparency, and cause the agent to behave in ways the user did not request.

Static analysis

No suspicious patterns detected.